Integrating Our Community

Year 1

Kay Snowley, Lara Edwards, Ben Crosby, Helen Tatlow

2 3

Introduction 4

Programme Team 5

Driver Programmes 6

Medicines in Acute and Chronic Care 7

Inammation and Immunity 9

Molecules to Health Records 11

Social and Environmental Determinants of Health 13

Big Data for Complex Disease 15

Infrastructure and Services 17

Technology Services Ecosystem 19

Federated Analytics 20

Trust and Transparency 21

Useable Data 23

Phenomics and Prognostic Atlas 24

Transforming Data for Trials 26

Capacity Building 27

UK Health Data Research Alliance 29

Regional Networks 30

Cambridge 31

London 32

Midlands 33

North 34

Oxford 35

Scotland 36

South West 37

Wales 38

Partnership Programmes 39

Hubs 40

Alleviate 41

DATAMIND 42

Discover-NOW 43

PIONEER 44

NHS DigiTrials 44

INSIGHT 45

Gut Reaction 45

DATA-CAN 46

BREATHE 47

BHF Data Science Centre 48

DARE UK 49

Contents

Introduction

Programme Team

Welcome to the rst HDR UK ‘Integrating Our Community’ pack! This pack is

intended to give you a snapshot of all the fantastic activities underway across

our community, signpost opportunities for wider involvement and highlight key

contacts for collaborations. It summarises the highly useful data provided in

the Year 1 workplans across Regions, Driver Programmes and Infrastructure

and Services Programmes. Furthermore, it highlights important Partnership

programmes such as the BHF Data Science Centre, Hubs and DARE UK,

promoting opportunities for integration, collaboration and cohesion. Strategies

and workplans that are currently still in development will be added to future

updates of this pack as they evolve, along with any new major developments

that oer future opportunities for integration and collaboration.

We would appreciate feedback on the pack, including what you would like to

see included in future versions (we expect to send out the next version in six

months’ time).

Ben Crosby - Programme Oicer

ben.crosby@hdruk.ac.uk

Lara Edwards - Programme Director,

Driver Programmes

lara.edwards@hdruk.ac.uk

Helen Tatlow - Programme Oicer

helen.tatlow@hdruk.ac.uk

Kay Snowley - Programme Director,

Infrastructure and Services

kay.snowley@hdruk.ac.uk

6 7

Medicines in Acute and Chronic Care

Key Activities in Year 1

• Develop a dra international framework for the identication of medicines-related harm.

• Prioritise which medicines research should be included in initial projects.

• Deliver a minimum viable product (MVP) for each use case, with tools to be rened in Year 2.

• Agree data specications for prioritised projects and data access processes/technologies.

• Develop a metadata catalogue for data, sharing this with the Innovation Gateway.

• Operationalise the identication of medicines-related harm, building automated tools for testing.

• Build the rst national medicines public stakeholder group, while co-developing patient and

public involvement and engagement (PPIE) strategy.

• Recruit PhDs and academic sta.

Driver Programmes

Workstreams

• WS1: Medicines innovation.

• WS2: Data and enabling technologies.

• WS3: PPIE.

• WS4: Capacity/capability building.

• WS5: Communications and stakeholder engagement.

• WS6: Operational delivery.

Wider Community

Linkages

• Regions: HDR North, Midlands, Wales, Scotland, South West.

• PIONEER Data Hub.

• Useable Data Pillar.

• Trusted Research Environments (TREs); Northern Ireland Honest Broker

Service (NI HBS), Secure Anonymised Information Linkage (SAIL) Databank,

Scottish National Data Safe Haven and regional TREs.

Dataset Priorities

• Primary care: SAIL Databank, Clinical Practice Research Datalink (CPRD).

• Secondary care including laboratory data: PIONEER, Scottish National and

Regional TREs.

• MHRA Yellow Card and FDA FAERS.

• Throughout Year 1 we will explore wider access to regional and national level

data via NHSE SDEs, OpenSAFELY, NI TRE and UK Biobank.

Technology and

Infrastructure

Priorities

• We will use SAIL and PIONEER environments from the outset, and with the

Technology Services Ecosystem (TSE) Pillar we will explore novel technology

services and solutions.

Driver

Programmes

8 9

Inammation and Immunity

Key Activities in Year 1

• Develop whole-system capacity to map the epidemiology, healthcare utilisation and outcomes for

common allergic and respiratory conditions for each of the UK nations in near-real-time.

• Use completed curated datasets for asthma, chronic obstructive pulmonary disease (COPD) and

interstitial lung disease (ILD) across all four nations, in order to provide UK-wide burden of

disease estimates.

• Extend recent work on the impact of respiratory illnesses on NHS winter pressures and a new

curated dataset for respiratory syncytial virus (RSV).

• Begin to advance mapping of variations in care processes and health outcomes.

• Begin to identify and develop novel near-real-time linkages with other suitable datasets.

• Recruit of ve more PhD students across the Programme.

• Create a strategy to develop, validate and test next-generation risk-prediction algorithms.

• Develop a PPIE strategy for the Programme.

• Extend engagement work to focus specically on policy inuence.

Driver Programmes

Key Opportunities

• Data Strategy: Developing a comprehensive, UK-wide medicines data strategy.

• Technology and Standards: With the TSE Pillar, testing novel

technology services.

• Streamlining Information Governance: Working closely with UK HDR Alliance

members to streamline information governance processes, developing best

practices and frameworks.

• Capacity building: Recruiting post-doctoral research assistants (PDRAs)

and PhDs.

• Policy and practice: Convening a stakeholder group.

• Levering funding and partnerships: Aiming to leverage 10-fold initial

investment through new partnerships and funding.

Key Challenges

• Loss of public trust in health data access.

• Sta/partner organisation recruitment.

• Challenges in accessing representative, longitudinal medicines data.

• Challenges with data linkage.

• A lack of involvement/engagement from diverse public and other

stakeholder groups.

• A lack of involvement from healthcare providers/practitioners.

How to get involved / PhD and training opportunities

• Non-clinical and clinical PhDs will be recruited across Year 1 and 2.

• PDRAs will be recruited throughout Year 1.

• Training programme (currently in development) that will co-design development plans for short

and longer-term career goals and how to achieve and resource them.

• Contact: Stephanie Robinson Larkin, S.Robinson-Larkin@liverpool.ac.uk.

Driver Programmes

Workstreams

• WS1: Develop whole-system capacity to map epidemiology, healthcare

utilisation and outcomes for common allergic and respiratory conditions.

• WS2: Develop novel near-real-time data linkages.

• WS3: Map variations in care processes and health outcomes and identify

opportunities for reducing inequalities.

• WS4: Develop, validate, and test - in large-scale, UK-wide clinical trials - next

generation risk-prediction algorithms.

• WS5: PPIE.

• WS6: Policy inuence and impact.

Wider Community

Linkages

• Driver Programmes: Molecules to Health Records, Social and Environmental

Determinants of Health, Medicines in Acute and Chronic Care and Big Data for

Complex Disease.

• BREATHE Hub.

• TSE Pillar.

• HDR Northern Ireland, HDR Wales, HDR North, HDR London, HDR Scotland.

Dataset Priorities

• Initial data sets and access: curated respiratory datasets, CPRD, DataLoch, NI

HBS, SAIL Databank.

• NHS England Secure Data Environment (NHS SDE) – in development.

• Link UK Severe Asthma Registry to primary care data.

• Priority data need: Regular, UK-wide GP data.

10 11

Driver Programmes

Technology and

Infrastructure

Priorities

• Whole UK (federated data) opportunities with linkage to key datasets in

socioeconomic / genomic / imaging data.

Key Opportunities

• Data strategy: key goal replication of four nation curated inammation and

immunity-related health data.

• Technology and standards: explore approaches to federation with Pillar 1,

DARE UK.

• Streamlining Information Governance: keen to work with Driver

Programmes and Pillars to explore solutions to regulatory hurdles.

• Policy and practice: Focus on mechanisms that can pivot to respond to

urgent policy needs.

• Capacity building: Second round of PhDs in Year 2, explore cross-Driver

Programme opportunities.

• Leveraging funding and partnerships: Continue to develop new partnerships

and opportunities.

Key Challenges

• Access to data across all four nations, including permissions for onward

sharing of curated datasets.

• Regular (near-real-time) primary care data updates in all four nations.

• Harmonising datasets: Consistency of methodology, coding, and data

denitions alongside close collaboration and communication between

analysts across the UK, to enable eective UK-wide analyses.

• Ensuring that PPIE is fully embedded; plans are in development that will

include signicant learning from other studies.

• Lack of team capacity to drive this work forward, either through skills,

sickness or leadership focus on HDR UK reporting and meetings.

How to get involved / PhD and training opportunities

• A second round of PhD opportunities is likely in Year 2; interested in collaboration opportunities

with other Driver Programmes.

• We are keen to discuss collaborations with other Driver Programmes on projects for early career

researcher (ECR) posts.

• Contact: Wendy Inglis Humphrey, winglis@ed.ac.uk.

Driver Programmes

Workstreams

• WS1: Population system genomics and Electronic Health Records (EHRs).

• WS2: Genomic medicine and EHRs.

• WS3: Molecular informatics tools and resources.

• WS4: Diverse and global cohorts.

Wider Community

Linkages

• British Heart Foundation (BHF) Data Science Centre.

• Big Data for Complex Disease Driver.

• Health Data Research Hubs.

• TSE.

Dataset Priorities

• Build on existing linkages to multi-omic cohorts to EHRs (NHS England, UK

Health Security Agency, UK HSA).

• Sub-licensing NHS England datasets to external researchers.

• Genomics England TRE data enrichment: Patient Reported Outcome

Measures (PROMS), administrative data, clinical imaging.

• International cohorts: Focus on obtaining collected data, generating

molecular data and outcomes follow-up.

Molecules to Health Records

Key Activities in Year 1

• Explore with the UK Longitudinal Linkage Collaboration (UK LLC) and TSE Pillar 2 the requirements

for multi-omic data TRE infrastructure development.

• Partner/explore partnering with UK-based multi-omic cohorts focused on non-European

ancestries.

• Evaluate mass spectrometry glycosylation assays deployed in several UK population cohorts.

• Enrich the Genomics England TRE data, focusing on specic data modalities – developing working

groups and expanding the HDR UK-Genomics England post-doctoral fellowship scheme.

• Co-develop a PPIE strategy and identify key public stakeholder groups for each activity.

• Explore partnerships with international cohorts to progress collaboration on informatics tools

and resources.

• Develop more diverse and global cohorts - begin to evaluate methods for linking cohort

participants to health records, and assess tools to ascertain causes of death reliably in existing

low-to-middle-income (LMIC) countries’ cohorts.

• Commence recruitment of Fellows and academic sta.

12 13

Driver Programmes

Technology and

Infrastructure

Priorities

• Currently no TREs are available nationally with the necessary technical

capabilities to enable eicient analysis of complex multi-omic EHR data. To

explore with TSE Pillar and UK LLC.

Key Opportunities

• Data strategy: Enrich longitudinal health-related data within the Genomics

England TRE, and multi-omics population cohorts.

• Data access: Make multi-omic data accessible via the Gateway, integrate with

the Cohort Discovery tool.

• Technology and standards: With the TSE Pillar and UK LLC, we will create

relevant methods and tools, such as federated analytics of multi-omics and

EHR data across TREs.

• Trust and transparency: Align to accelerate trustworthy data access, and

PPIE strategy.

• Partnerships: Build on major collaborations, such as UK Biobank, NIHR

BioResource and build on major international collaborations such as the

Global Alliance for Genomics and Health.

• Leveraging funding and partnerships: Aim to leverage initial investment with

external funding; and engaging multiple industry partners.

Key Challenges

• Not achieving priority EHR data linkages due to application delays or changes

in governance.

• Lack of a Secure Computing Environment required for data analysis of

complex multi-omic data alongside EHR data.

• Accessing representative, longitudinal medicines data.

• Data linkage issues.

• A lack of involvement/engagement from diverse public and other

stakeholder groups.

• A lack of involvement from healthcare providers/practitioners.

How to get involved / PhD and training opportunities

• We plan to extend a joint initiative with Genomics England’s Clinical Interpretation Partnership

(GeCIP) to establish a cadre of clinical and non-clinical Fellows in Year 1 to support ongoing GeCIP

rare disease research.

• We plan to recruit a Clinical Informatician in the second quarter of Year 1.

Contact: Richard Houghton, rh12@sanger.ac.uk.

Social and Environmental Determinants of Health

Key Activities in Year 1

• Develop governance principles for sharing Unique Property Reference Numbers (UPRN) across

TREs, avoiding re-identication in attribute data.

• Carry out a scoping review of literature on methods for protecting privacy.

• Advance best practice technical principles for importing large Geographic Information System

(GIS) data across assets/TREs.

• PPIE activity on geoprivacy issues, UPRN and health data linkage.

• Achieve permissions to link to UPRN data in Imperial College London’s Small Area Health Statistics

Unit (SAHSU) (Workstream 2).

• Achieve permissions for Education and Child Health Insights from Linked Data (ECHILD) cohort

and the UCL Kids’ Environment and Health Cohort (KEHC) to link to UPRN.

• Achieve support and permissions for enhancing and linking historic cohorts – UK LLC cohorts with

UPRN, Scottish cohort analysis of health outcomes, mortality and UV/sunlight exposure.

• Begin PhD and sta recruitment.

Driver Programmes

Workstreams

• WS1: Infrastructure, data, methods and governance.

• WS2: Heath in Aging Populations.

• WS3: Health and Development from birth to adulthood.

• WS4: Linkage to historic cohorts.

Wider Community

Linkages

• Molecules to Health Records, Inammation and Immunity, and Medicines in

Acute and Chronic Care Driver Programmes.

• UK Longitudinal Linkage Collaboration (UK LLC).

• CLS national cohort studies.

• HDR Scotland, HDR Wales, SAIL Databank.

• Pillars: Trust and Transparency, TSE.

• HDR Alliance.

Dataset Priorities

• SAHSU Research Database.

• ECHILD.

• KEHC Research Database.

• UK LLC.

• Scottish historic Population Platform.

• SAIL Databank.

14 15

Driver Programmes

How to get involved / PhD and training opportunities

• Two three-year research posts will be recruited at Swansea in Autumn 2023, providing the

opportunity to pool resources across Driver Programmes to make ve-year posts.

• Contact: Matthew Lilliman, m.lilliman@ucl.ac.uk.

Big Data for Complex Disease

Key Activities in Year 1

• Awarded six Big Data for Complex Disease (BDCD) PhD projects and studentships across the HDR

UK community (ve have been recruited, one will be readvertised).

• We are preparing to launch a BDCD fellowship competition led by Queens University Belfast with

the aim of launching in September 2023.

• Enhance disease-based cohorts: A data linkage platform is under development.

• Expand our set of reproducible data curation and analysis pipelines, including phenotype

algorithms.

• Discuss solutions to share/merge environments, data access and analysis, sharing of code and

PPIE.

• Recruit key posts (eg, programme manager).

Driver Programmes

Technology and

Infrastructure

Priorities

• Explore TRE capability for linking, generating and analysing large-scale

environmental exposure data, and federated approaches – such as

collaborating with the Natural Environment Research Council (NERC) Digital

Solutions Programme, linking in with the TSE Pillar.

Key Opportunities

• Data Strategy: We are keen to explore cross-links in Workstream 4 with other

Driver Programmes on the association between environment and health in

older adults (respiratory, cardiovascular disease outcomes).

• Data access: Widening access to SAHSU dataset.

• Technology and Standards: We are keen to work with the Infrastructure and

Services Pillar and HDR Alliance to inform harmonisation of methods for

linking to place across TREs.

• Trust and Transparency: We will work with the HDR Alliance on governance

aspects and public support for geospatial data linkage.

• Partnerships: Data custodians of health and health-related data, recruited

cohorts and surveys, organisations that generate environmental data,

TREs, regulators and policy bodies. (eg Oice for National Statistics (ONS),

Ordnance Survey, UKHSA, Geospatial Commission, NHS England, SAIL

Databank, the Met Oice, NERC).

• Levering funding and partnerships: Priority is to leverage sustainable

funding for ECHILD.

Key Challenges

• Public support: Risk that privacy groups and the wider public will not support

the work.

• Governance: Risk that the Health Research Authority will not approve the

health data assets in the scope of this programme to hold UPRN.

• Data providers: Risk that data providers (e.g. NHS England and ONS) do not

provide UPRN to the health data assets within the required timescales.

• TREs: Risk that changes to TREs could delay deliverables.

• Progress against the relevant novel deliverables.

• Risk of delay with sta and student recruitment.

Workstreams

• WS1: Better predict diseases such as cancers and cardiovascular

diseases, thereby improving screening, detection, early diagnosis

and prevention strategies.

• WS2: Better understanding of the inter-relationship between these

complex diseases.

• WS3: Better understanding the impact of inequalities in personal and

system-level characteristics and national/ regional geography in order to

inuence and mitigate the negative impacts on incidence and outcomes

of these conditions.

• WS4: Capacity- and capability-building.

• WS5: PPIE, Stakeholder and Policy Engagement.

• WS5: Operational delivery, Data and Infrastructure.

Wider Community

Linkages

• BHF Data Science Centre, DATA CAN Hub.

• Molecules to Health Records, Inammation and Immunity, and Medicines in

Acute and Chronic Care Driver Programmes.

• UK LLC.

• HDR Wales (SAIL Databank), HDR Scotland, HDR South West.

Dataset Priorities

• National Institute for Cardiovascular Outcomes Research (NICOR) – broader

linkage and access.

• NHS England SDE: CVD-COVID-UK/COVID-IMPACT, DATACAN. Priority to

enhance DATACAN Data linkage – and cancer data completeness and

provenance (eg Cancer Outcomes and Services Dataset (COSD)).

• Devolved nations’ TREs via CVD-COVID-UK/COVID-IMPACT Consortium.

• Virtual Cardio-Oncology Research Initiative (VICORI).

• Social data: Current opportunity in Wales, in time there will also be

opportunities in sub-national NHS England SDEs.

16 17

Technology and

Infrastructure

Priorities

• Broader and wider linkage to NHSE datasets currently available in DATACAN

and CVD-COVID-UK/COVID-IMPACT.

• Enhancing disease-based cohorts: Data linkage platform under development

with SAIL Databank to hold cohort study data linked to health data.

Key Opportunities

Data Strategy: we are keen to identify opportunities to collaborate, eg there

is lots of data linkage potential across the Driver Programmes, overlapping

ambitions with others such as Inammation and Immunity.

• Data access: We are keen to enable cross-Driver Programme access to CVD-

COVID-UK/COVID-IMPACT, DATACAN.

• Technology and Standards: Work with Infrastructure and Services

Programmes.

Key Challenges

• Reorganization could impact capacity and headspace at NHS England to

make rapid progress with additional datasets, linkages and data curation

which are required.

• Risk of delay with sta and Fellow recruitment.

How to get involved / PhD and training opportunities

• The Fellowship programme will launch in September 2023, providing opportunities for cross-

Driver Programme collaboration/projects.

Contact: Cathie Sudlow, cathie.sudlow@hdruk.ac.uk and Helen Tatlow, helen.tatlow@hdruk.ac.uk.

Driver Programmes

Infrastructure

and Services

18 19

Infrastructure and Services

Technology

Services

Ecosystem

FAIR access to population-scale data at depth and breadth is needed to enable

linkage of data from many custodians, and federated analyses are needed

across SDEs for many health data researchers across the UK and globally. This

Pillar aims to achieve this by bringing together a UK-wide team of leading

technologists, data scientists from across academia, SDE providers, industry

and the NHS, all of whom are committed to the assembly of an ecosystem of

services. Embedding a collaborative, federated delivery model will enable

greater patient and public benet than any single organisation can achieve in

isolation, whilst still maintaining autonomy of all involved.

Trust and

Transparency

To deliver HDR UK’s mission to enable data-driven research that improves

people’s lives, patients and the public need to have trust and condence in the

safe, secure and trustworthy access to, and use of, their data. Demonstrating

trustworthiness and building public condence on a national scale in a complex

data landscape is challenging, but it is vital in order for the UK to achieve

its research potential. This is particularly crucial in health data research. To

support this mission, this Pillar underlines HDR UK’s commitment to leading

both meaningful involvement and engagement with the public, and a robust,

transparent, trustworthy, and streamlined governance and ethics framework.

Useable Data

The aim across the Useable Data Pillar is to support researchers to identify

datasets that meet their needs, and to minimise the amount of pre-processing

or curation work they must do to make it ‘t for purpose’. The work will see the

building of reuseable, open, and extensible soware infrastructure through

the ‘Data Standards’, ‘Phenomics and Prognostic Atlas’ and ‘Transforming Data

for Trials’ Infrastructure Programme Workstreams, which will provide support

across the data-to-analysis pipeline. Together, these workstreams focus on

alignment of approaches to data and metadata (including phenotypes), with the

aim of developing and driving adoption of consistent standards and formats.

Capacity Building

The Capacity Building Pillar will implement a unique range of health data

training resources informed by the cutting-edge science and technology

developed by our Infrastructure and Services and Driver Programmes, which

will advance a new talent pool that can apply advanced health technologies and

deploy those advances within the ecosystem. By accelerating the early adoption

of emerging data skills, it is envisioned that HDR UK can make transformational

impacts in health data science training content, research practice, training

delivery approaches, and outreach and connections to dierent communities.

Technology Services Ecosystem

Key Activities

• Complete a plan for community training materials.

• Launch Technology Strategy paper and webpage summary.

• Gateway mark 1 placed in maintenance mode, Gateway Mark 2 launched.

• Complete beta testing for push and pull Fast Memory Access approach.

• Promote of cohort discovery tool to potential data sources.

• Promote and train tools to support mapping to OMOP from metadata and generate Extract,

Transform and Load (ETL) scripts across the community.

• Undertake a consensus approach to develop priority data platform requirements and prepare

proposals to respond to suitable funding calls.

Infrastructure and Services

Wider Community

Linkages

• HDR UK Tech Community Principles developed and implemented.

• New collaborations identied for creating/interoperating with a new piece of

technology within the ecosystem.

• Established requirements gathering feedback process, which is documented,

embedded and sharable.

• Soware development principles reviewed and agreed by the community.

• Options appraisal documented (and shared) for the use of Observational

Medical Outcomes Partnership (OMOP) mapping to support enhanced

Gateway search capability at a eld level Tools to support mapping to OMOP

from metadata and the generation of ETL scripts across the community have

been promoted and trained.

Key Opportunities

• Implementation of HDR’s Technician Commitment.

• Completion of testing and rollout of pre-production system for Mark 2.

• Promotion of the use of OMOP as a data standard across the Alliance and

collaboration with the European Health Data Evidence Network (EHDEN) to

fund groups to map to OMOP, plus development of UK Observational Health

Data Sciences and Informatics (OHDSI) node.

Key Challenges

• The re-engineering of the Gateway could take longer than a year and the

solution developed does not meet needs of community.

• Re-writing and duplication of solutions/services.

• Data governance amendments not approved Risk that data custodians will

not see the value of onboarding their data. Risk of a ‘scatter gun’ approach in

attempting to meet multiple use cases.

• Trying to do too much across what are currently disparate activities and

plans, and a risk of multiple stakeholders engaging in diering priorities.

20 21

Infrastructure and Services

Wider Community

Linkages

• Transition programme aligning DARE UK into core programme.

• Opportunities for international collaboration assessed.

• Community established across HDR UK for Federated Analytics.

• Individuals from across the Driver Programmes identied to be part of a

wider network.

Key Opportunities

• Prototype of HUTCH Five Safes RO-Crates developed.

• Roadmap for international engagement established.

Key Challenges

• Driver Programmes may develop their own solutions.

• Balancing international opportunities with national delivery.

• Uncoordinated interest from industry.

Trust and Transparency

Key Activities

• Work towards building streamlined data access information governance models through the Pan-UK

Data Governance Steering Group.

• Work towards data custodians adopting a single core set of questions that need to be submitted by

researchers for access to data. This objective will be necessary to achieve the vision set out in the

Department of Health and Social Care data access policy published in June.

• Create a set of core principles for data access in TREs, including identication of customisable

controls relevant to some TREs - the principles will be developed following a benchmarking activity

involving review and comparison of data access agreements for established TREs and will be mapped

against the Five Safes Framework.

• Collaborate with Public Advisory Board (PAB) members to create Transparency Standards to be

adopted by the Alliance and activities to promote adoption.

• Creation of a standardised Data Depositing Agreement (DDA) to streamline the contractual process

between data contributors and TRE host organisations.

• Information Governance management for the set-up of the BHF DSC cardiovascular disease and

diabetes cohorts TRE.

• Promote inclusivity in PPIE activities by attending three festivals that target under-represented

groups over the next 18 months and evaluate the impact.

• Engage with key stakeholders to identify community groups to engage in health data-related

activities to increase representation.

• Create a community that will focus on the legal challenges of international data sharing.

• Work with public members and PPIE professionals to update the HDR UK PPIE strategy for the second

ve years.

Infrastructure and Services

Wider Community

Linkages

• Create a single map of all approvals required to gain access to data; an Action

Force has been set up to deliver this piece of work.

• Map roles and responsibilities for all parties involved in data access in TREs

against UK General Data Protection Regulation (GDPR) to add clarity and

consistency across contractual arrangements and obligations. This will then be

turned into a functional tool to aid in division of responsibilities and to signpost

which contracts will be required.

• Guidance for use of the Data Access Agreement (DAA) templates will be

developed including review and input from public members.

• Pilot at least one community engagement event and evaluate impact in the next

18 months.

• Launch pilot campaign targeting focus group regions.

• Collaborate with PAB members and stakeholders interested in advancing PPIE

practices, such as ‘Public Engagement in Data Research Initiative’ and ‘Shared

Commitment to Public Involvement’ partners.

Federated Analytics

Key Activities

• Initial PhDs oered to candidates.

• Initial Driver Programme requirements gathered.

• SAIL and PIONEER to utilise TRE-FX (an application stack for Federated Activities being developed

for TREs or SDEs) with Five Safes RO-Crates in a production environment.

• TRE-FX with Five Safes RO-Crates, though co-development with BC Platforms able to be a

production alternative to BC Link.

How to get involved

• Contact: Emily Jeerson, emily.jeerson@hdruk.ac.uk.

How to get involved

•

Contact: Phil Quinlan, philip.quinlan@nottingham.ac.uk and Carole Goble, carole.goble@

manchester.ac.uk.

22 23

Infrastructure and Services

Useable Data

Key Activities

• Maximise benets from UK implementation of EHDEN partnerships:

• Publish a review of the current landscape and survey results.

• Increase awareness and adoption of recommendations for collection of ethnicity data.

• Map the landscape of international data standards bodies.

• Identify data priorities, informing data strategy.

• Support improvements of data quality and completeness of diversity-related data in NHS settings

(in alignment with Alliance activities).

Wider Community

Linkages

• Establish an OMOP CDM Special Interest Group, as a sub-group of the Alliance

Data Oicer Group (Inaugural meeting in the rst quarter, two additional

meetings planned throughout the year).

• Engage new partners including NHSE Health and Disparity Unit; Race and

health Observatory; Understanding Patient Data; Wellcome.

• Convening the Data Oicer Group, a community forum to discuss data

standards, data quality, metadata, ontology and terminology.

Key Opportunities

• Map the landscape: understand and analyse the current level of adoption of

the OMOP Common Data Model including desk-based research, individual

interviews and a survey.

• Support data strategy integrated objectives.

• Identify data standards priorities with input from Driver Programme, UK and

international investments, Alliance partners, HDR UK regions.

How to get involved

• Contact: Monica Jones, m.c.m.jones@leeds.ac.uk, and Alex Knight, alex.knight@hdruk.ac.uk.

Infrastructure and Services

Key Opportunities

• Promote standardisation of data access governance processes.

• Streamline data access contracting.

• Increase the number of HDR UK Voices members by 100% (203) in the next year

• Launch Public Engagement in Data Research Initiative (PEDRI) pilot campaign

and engage at least 500 members of the public.

• PAB can actively contribute to improving transparency in data use, enhancing

public trust in health data.

• Partner with organisations and initiatives focused on PPIE practices brings

access to a wealth of expertise and resources in the eld.

• Broaden the participation of under-represented groups in health data research

and eorts.

• Incorporate diverse perspectives and expertise in the new HDR UK PPIE strategy.

Key Challenges

• There can be confusion and lack of consensus around ‘Data Controllership’ and

many parties may be Controllers in the process of data access in TREs. For this

reason, we will dene responsibilities by the following roles: Data Contributor,

Host Organisation, TRE Platform, TRE Service Provider, User Organisation,

Approved Researcher.

• Established TREs/SDEs and data custodians may be reluctant to change

processes in place due to operational/resource impacts and/or lack of

agreement or understanding of the benets of standardisation.

• Maintaining trust in data use while promoting access to health data can be a

delicate balance, with concerns about privacy and security.

• Stakeholders involved in advancing PPIE practices may have diverse interests,

priorities, and goals. Balancing these interests and aligning them with HDR UK’s

objectives can be challenging.

• Establishing clear metrics and methodologies for assessing the success of the

pilot activities may be diicult.

• Failing to identify the right groups can result in ineective outreach and a lack

of meaningful participation.

• Striking a balance between the input from the public and the PPIE professionals

involved in the development of the new HDR UK PPIE strategy.

How to get involved

• Contact: Cassie Smith, cassie.smith@hdruk.ac.uk.

24 25

Infrastructure and ServicesInfrastructure and Services

Phenomics and Prognostic Atlas

Wider Community

Linkages

• Develop the governance to assemble and reward a single national productive,

interdisciplinary team to deliver common goals of PPA, as funded by HDR

UK; seek synergies across HDR UK including with Data Standards, Trials, BHF

DSC, Big Data for Complex Disease, and the pre-existing Atlas and CogStack/

Foresight teams. We will seek to identify clear relationships, communications

and lines of accountability, noting that these teams comprise people funded

from dierent routes within and outside of HDR UK.

• Prototype development of soware services/components with open APIs

e.g. results service, clinical guidelines, papers service, people service,

publications service, ontology service, DOI service (nal quarter).

Key Opportunities

• Horizon scanning and planning how to attract additional funding and take

advantage of the opportunities presented by recent advancements in LLMs.

• Dra of a series of clinical specialty papers for publication arising from

Phenomics and Prognostic Atlas (PPA), eg given our close dialogue with Big

Data for Complex Disease, an initial focus might be in cardiovascular and

oncology specialties.

• Begin to systematically add phenotypes with molecular markers, focusing on

the requirements of clinical trials (nal quarter).

• Formalise core Atlas ontology, including specialties, into an international

standard such as the Open Biological and Biomedical Ontologies (OBO) or

similar format (third quarter).

Key Activities

• Develop, validate and deliver the Disease Atlas and disease phenotypes as a new body of methods

and knowledge built at nationwide scale to inform research across HDR UK, clinical practice and

health.

• Advance disease phenomics and the Disease Atlas through (a) hospital-wide approaches using

unstructured and structured data to generate disease phenotypes in all patients and (b) using

large language models (LLMs) trained in nationwide data (initially structured).

• Seek opportunities for sustainable growth of products and services from Atlas, the Phenotype

Library and Foresight, through external funding and commercialisation.

• Prototype initial demonstration of value / user journeys on the Disease Atlas interactive website.

• Develop an enhanced version of Phenotype Library, holding PPA content and integrations with the

PPA website.

Key Challenges

• Evolving PPA in the light of developments in LLMs: The risk is that we do not

dynamically update and prioritise the aims and direction of the theme, in the

light of national and international advances - eg rapid developments in the

last few months in generalised medical AI may supersede, or challenge, some

elements of the submitted proposal (large language models (LLM) are not

mentioned at all in the QQ2 submission) Mitigation: establish a PPA Board,

tasked with dynamically updating the priorities of the theme within the

overall framework.

• Learning from related national and international initiatives: the risk is that we

do not learn from diverse related national initiatives (eg Open Prescribing)

and international initiatives (eg Global Burden of Disease). Mitigation: the

PPA Board will use evidence comparing initiatives to inform decisions about

priorities for the team’s time.

How to get involved

• Contact: Harry Hemingway, h.hemingway@ucl.ac.uk.

26 27

Infrastructure and Services

Capacity Building

Key Activities

• Growing the health data learning curriculum through HDR UK’s Futures Learning Platform:

This content builds on the HDR UK Futures platform that has benetted from the wide support

of the HDR UK community. We are shaping the design and delivery of innovative new training

programmes to accelerate learning, build new capabilities and promote the sharing of best

practice across Alliance partners.

• Technician Commitment: In 2022 we made the decision to sign up to the Technician’s

Commitment and published our action plan in June 2023 in which we described the actions we

would take to support technicians both internally and externally. We will be deploying this action

plan from September 2023 onwards.

• HDR UK Turing/Wellcome PhD programme: Our agship four-year PhD programme aims to train

the future leaders in health data science. We have recruited four of the ve cohorts with the nal

cohort due to be recruited for an October 2024 start date. Cohort four will start their training year

in October 2023.

Infrastructure and Services

Transforming Data for Trials

Key Activities

• The Trials Stakeholder Prioritisation Forum, working with the Alliance, will establish a forum which

brings together the key stakeholders in a four-nation approach, including data custodians (eg NHS

England), users of data in the trials community (eg trial teams) and users and interpreters of the

ndings of trials (eg the Medicines and Healthcare Regulatory Authority, MHRA).

• PPIE: this work-stream contains two elements: (1) patients and the public will be involved in

developing the strategic aims and planned output of the Transforming Data for Trials program; (2)

the program will identify or develop PPIE output relating to the use of healthcare systems data in

clinical trials (these outputs will be accessible via the route-map).

• A thematically curated catalogue of case studies will be developed from the spectrum of trials

from across the UK and international trials, if relevant.

Wider Community

Linkages

• Training resources will be developed, aiming to enable trial teams to use

healthcare systems data.

• Data utility comparisons are methodological research studies in which the

utility of study outcomes identied from healthcare systems data is assessed

compared to outcomes ascertained through the traditional ‘gold standard’

trial methods.

Key Opportunities

• Route-map for trials using healthcare systems data: The ‘route-map’ for

healthcare systems data which will be parallel and complement the existing,

widely accessed NIHR Clinical Trial Toolkit route-map.

• Knowledge transfer, exchange and mentoring -this work-stream will establish

eective methods knowledge transfer between institutions in relation to use

of healthcare systems data.

• Demonstrating the data integrity and provenance of healthcare systems data

for clinical trials.

Key Challenges

• Timely appointment to required posts.

• Duplication of content with other organisations working in this space.

• Ability to deliver required training (funded across ve years of the

programme) while supporting release (eg aer 18-24 months) of the matched

route-map.

• Failure to engage with the key relevant stakeholders.

• Insuicient sta and issues in appointing people to posts.

• Data providers don’t engage.

How to get involved

• Contact: Marion Mafham, marion.mafham@ndph.ox.ac.uk and Matt Sydes,

matt.sydes@hdruk.ac.uk.

28 29

Infrastructure and Services

UK Health Data Research Alliance

Key Activities

• Become a recognised and inuential network of partners to enable trustworthy use of health-

related data for research and innovation through four main 12-month priorities:

• Sustain partnership with NHS England to inform and maintain alignment with secure data

infrastructure developments and structural changes.

• Maximise benets from the UK implementation of the EHDEN partnership.

• Maintain and extend alignment with other key national organisations e.g. ONS, Administrative

Data Research UK, Alan Turing Institute, Association of Medical Research Charities (AMRC),

Association of the British Pharmaceutical Industry (ABPI) and the Bioindustry Association (BIA).

• Extend Alliance membership to non-data custodian organisations.

Wider Community

Linkages

• Grow and strengthen the Alliance to include major health-relevant data

controllers, industry and other representative bodies.

• Support development, implementation and adoption of recognised

standards, best practice and tools to accelerate trustworthy access to, and

use of, data.

• Develop and implement a communication strategy to improve engagement,

diversity and growth.

• Taking on a convening role, we will develop and implement principles

and best practice tools and policies in various streams including trust and

transparency, technology and data.

• Encourage recognition and adoption of standards (eg data use registers, TRE

principles and the Five Safes Framework) in the UK and beyond.

Key Opportunities

•

Enable Alliance members to contribute towards a UK-wide strategy to

accelerate access to, and use of, data.

•

Maintain and extend alignment with other key national organisations

including AMRC, BIA and ABPI.

•

Sustain partnership with NHS England to inform and maintain alignment with

secure data infrastructure developments and structural changes.

How to get involved

• Contact: Paola Quattroni, Paola.Quattroni@hdruk.ac.uk.

Infrastructure and Services

How to get involved

• Contact: Sarah Cadman, sarah.cadman@hdruk.ac.uk

Wider Community

Linkages

• Developing diverse pathways in health data science.

• Supporting Driver Programmes, the Alliance and Infrastructure and Services:

Training champions will take up secondment positions lasting between

three and 12 months, which will be embedded within Driver Programmes

and Infrastructure and Services. They will identify training opportunities and

requirements both within individual and across programmes and produce

cutting-edge training tools to address these requirements.

• Cross-sector transition grants: The Capacity Building Pillar will seed-fund up

to 12 cross-sector transition grants to enable PhD students, HDR UK Fellows or

industry-based collaborators to undertake three to six-month placements in

dierent sectors (eg, academia to industry or vice-versa).

• Masters Scholarships: HDR UK is partnering with charities and industry to oer

master’s degree scholarships to quantitative students.

• Medical School Data Science Survey: We have issued a survey to medical

schools across the UK in conjunction with the Academy of Medical Sciences

and NHS England. We have recruited an intern from the Black Internship

Programme to develop the analysis for this survey during the summer of 2023.

Key Opportunities

• Supporting future health data science leaders.

• Black Internship Programme – The target is to recruit 750 interns over ve

years; 96 were recruited in 2022/24 for the Black Internship Programme and

this number continues to grow year on year. We secured external funding

for eight times more internships on this programme (via the EPSRC Digital

Health Hubs).

• Alumni Network – Our target was to have 5,000 members over ve years and

to date we have recruited over 350 members from core programmes (fellows,

PhD students, Masters students and interns). Furthermore, network leads have

been established and a programme of career building activity is underway.

30 31

Regional Networks

Regional Networks are designed to leverage expertise and forge partnerships across the regions. The

geographical representation of the Network continues to be invaluable. It enables vital contributions

from leading researchers and research organisations to focused, UK-wide research programmes, and

opens up regional capabilities in infrastructure, data, services, the NHS, PPIE and innovation.

Each of the Regional Networks has:

• Clear leadership, which continues to build partnerships, activities and innovation across the respective

regions as well as an overarching ambition to improve national capabilities.

• Well-established links with colleagues, including key local data custodians, government, the NHS and

public health, localised specialist partnerships, and skills across the four nations.

The substantial dierences between neighbouring nations, and regions within a nation, imply complex

interactions between biological (eg genetic), environmental and social factors. Working with health data

science systems in these regions will be essential if the UK is to address these inequalities.

Cambridge

Wider Community

Linkages

• A monthly Cambridge-hosted multi-disciplinary seminar series, which is open

to the entire community.

• We mobilise data resources at the regional level to complement national

resources, and ensure connectivity with cognate activities, locally and

nationally: the BHF Data Science Centre, EHDEN, Cancer Data Driven

Detection (CD3) initiative and the East of England sub-national SDE.

• We support the University of Cambridge’s health data science master’s

programme and PhD studentships to be hosted at Cambridge will be funded

through the Molecules to Health Records Driver Programme.

Key Opportunities

• Shape development of the East of England sub-national SDE.

• Connectivity across the Cambridge ecosystem to ensure alignment on TRE

infrastructure.

Key Challenges

• Complexity and risks of accessing and mobilising data from local /

national sources.

• Risk of fragmentation of data assets across dierent TRE infrastructure, which

could lead to diiculties in data access or limit analyses.

Key Activities

• Industry engagement with pharmaceutical companies (eg AstraZeneca, GSK) to map out areas of

potential synergy on which to collaborate via our PhD programmes.

• Embed PPIE in the full life cycle of research by engaging with the diverse participants of Cambridge-

led research cohorts and seeking their input on a wide range of documents and workshops.

How to get involved

• Join our monthly multi-disciplinary seminar series and help spread the word for our 2024 intake of

HDR UK-funded PhD students.

Contact: Richard Houghton, rh12@sanger.ac.uk.

Regional

Networks

32 33

Wider Community

Linkages

• Forge partnerships across regions to answer fundamental questions together

and deliver better research and care through inter-regional engagement

activities and workshops.

• Incorporate novel inter-regional insights from analyses already underway in

national TREs to help prioritise pilots, specify knowledge gaps and add value

to whole hospital approaches.

Key Opportunities

• Focus on whole hospital data approaches as a complementary eort to

other regions.

• Highlight benets of enabling access for researchers to structured and

unstructured data in whole hospital systems.

• Identify partner hospitals which have permissions and capabilities for whole

hospital analyses, to demonstrate what is possible.

Key Challenges

• Risk of access to hospital data being withdrawn or de-prioritised, possibly

mitigated by inter-regional prioritisation of research areas of value to the NHS.

• Comparisons of hospitals’ data may instigate unnecessary feeling of

competition, which carries risks of negative views.

Regional Networks

London

Key Activities

• Inter-regional development of methods and applications of ‘whole hospital’ health data research to

inform clinical practice, health policy, and research.

• Develop specic ‘methods pilots’ to generate inter-regional proofs of concept for use as building

blocks of wider value across the HDR community; in addition, stimulate and incubate pathnder

projects to understand the implications of using whole-system approaches for research, clinical

practice, and policy.

How to get involved

• Contact: Harry Hemingway, h.hemingway@ucl.ac.uk, and, for inter-regional capabilities and whole

hospital approaches to health data research, Sinead Langan, sinead.langan@lshtm.ac.uk.

Regional Networks

Midlands

Key Activities

• Scale of health data research activity across Midlands requires a mapping exercise to understand

all Health Data Science Projects.

• The rst quarterly insight sharing day is planned for September 2023.

• We are developing a PPIE strategy to ensure that patients and the public are involved in our work.

Wider Community

Linkages

• Maximising opportunities for our region through the HDR UK Midlands

Regional Community Platform.

• Develop a training strategy to upskill the region.

• Work closely with other HDR UK regions to maximise opportunities.

• Connect our region with other components of HDR UK, such as Driver

Programmes and Infrastructure and Services.

Key Opportunities

• Upskilling and improving the knowledge and capabilities of our workforce by

learning from each other.

• Continue to grow our membership base.

Key Challenges

• Key regional groups, networks and organisations are not involved in

our work.

• Impact, outputs and outcomes not fully understood or reported.

How to get involved

• Contact: Kevin Dunn, k.w.dunn@bham.ac.uk.

34 35

Wider Community

Linkages

• Leverage further funding by submitting collaborative regional applications

and build research capacity by supporting doctoral / postdoctoral fellowship

applications - the focus of applications will be reective of the research

expertise of the North, and DHSC priorities for improving health outcomes in

health and social care.

• Work with other HDR UK regional networks on a shared purpose, to enable

access to secondary care data, host knowledge exchange activities to

encourage this work, and leverage further funding.

Key Opportunities

• Shape and support NHS and local authority workforce analytical capabilities

within the region through co-developed knowledge exchange activities,

including joint workshops and seminars.

Key Challenges

• Potential capacity constraints on the North’s Associate Directors ability to

submit additional funding applications owing to clinical duties.

• Stakeholders fail to realise synergy and value of integrating with

the programme.

Regional Networks

Wider Community

Linkages

• Remain highly engaged with NHS England and ensure the emerging new

system is t for purpose and much more eicient than previous iterations.

• Recruit to a variety of schemes for internships such as UNIQ+ and the HDR UK

black internship programme.

• Initiate a programme of seminars and workshops to share knowledge and

expertise with the health data research community.

Key Opportunities

• Establish new and enhance existing communication channels to ensure all

working in health data science in our region are aware of opportunities for

collaborative work.

• Increase awareness, and therefore, number of users in all health data

research secure data environments.

• Ensure strong patient and public engagement throughout our programme

of work.

• Improve the student experience and success rates.

• Early, mid and senior researcher applications for research funding and

success rates.

Oxford

Key Activities

• Strengthen the HDR network across the region and beyond, by building stronger links with existing

health data research groups and establishing new collaborations to increase eiciency and

translational impact.

• Build on existing initiatives to expand the HDR infrastructure and improve secure access for

approved research data that also protects the public interest.

• Continue building a strong health data science workforce by identifying and fostering new talent in

internships and PhD students and supporting existing sta with their career aspirations.

Regional Networks

How to get involved

• Contact: Eva Morris, eva.morris@ndph.ox.ac.uk, and Cecilia Lindgren, cecilia.lindgren@bdi.ox.ac.uk.

North

Key Activities

• Ensure new discoveries using health data translate into improvements in health and wellbeing, by

establishing strong links with our Integrated Care Systems (ICS), SDEs, TREs, NHS leaders, public

health bodies, data custodians, charities, companies and academic groups engaged in health data

research.

• To catalogue TREs, SDEs, and Federated Data Platforms (FDPs) across the north of England to

provide an overall view of these assets across the region, and to facilitate data sharing across

multiple systems, linking to the HDR UK Technology Services Ecosystem (Pillar One Programme).

How to get involved

• Contact: Steph Robinson-Larkin, s.robinson-larkin@liverpool.ac.uk.

36 37

Regional NetworksRegional Networks

South West

How to get involved

• Contact: Serena Tricarico, s.tricarico@ed.ac.uk.

Wider Community

Linkages

• Launch of the Advanced Computer Science Summit, hosted in Edinburgh

under the stewardship of HDR UK: The aim is to encourage those in our HDR

networks who have an interest in cutting-edge applications of computing

to consider the fast-developing wave of new techniques, while forging new

relationships (including industry) that could help us inhabit this new space.

Key Opportunities

• Connecting with the ve DARE UK Driver Projects and determining how HDR

UK Scotland can work with them.

Key Challenges

• Failure of key stakeholders to engage in our data management work and

Advanced Computer Science Summit.

• Dependencies on the TSE as a separate project with its own timescales may

impact our ability to deliver across the region.

Wider Community

Linkages

• Leadership roles in Big Data for Complex Disease and Medicines in Acute and

Chronic Care Driver Programmes.

• BHF Data Science Centre.

• SAIL Databank /UK SeRP (HDR Wales) hosting UK LLC.

• Work closely with other HDR UK regions to maximise opportunities.

Key Opportunities

• Bristol-based post and PhD to support Medicines in Acute and Chronic Care

and Big Data for Complex Disease Driver Programmes.

• Development and wider access to UK LLC.

• Year 1 South West regional capacity building events.

Key Challenges

• Reduced spending power.

• Potential for delay to GW SDE outside regional control.

• Potential delay to PhD/sta recruitment.

Scotland

Key Activities

• Engage with Research Data Scotland to increase the diversity of accessible data at high quality to

researchers; reduce latency in data workows; introduce appropriate, responsible automation into

data cleaning and analytics; and collaborate on cross-UK data standardisation.

• Encourage cross-working with the development of the HDR UK TSE (Infrastructure Pillar 1) to

propagate best practice across the health data research network.

Key Activities

• Leadership within and development of the Great Western (GW) sub-national SDE.

• Continued role in development of the NHSE national SDE and OpenSAFELY.

• Training and capacity building in health data in South West region- new Member Plymouth

University to co-lead.

• Leadership and continued development of the UK LLC.

How to get involved

Contact: Lizzie Huntley, lizzie.huntley@bristol.ac.uk.

38 39

Wales

Key Activities

• Create and maximise opportunities, dissemination and outreach to further build an agile

community and PPIE.

• Work with the Driver Programmes includes: Medicines in Acute and Chronic Care, Social and

Environmental Determinants of Health, Inammation and Immunity, Phenomics.

Regional Networks

How to get involved

• Contact: Alysha Morgan, a.l.morgan@swansea.ac.uk.

Wider Community

Linkages

• Dementia Platform - Building on the dementia platform for grants in the

Medicines in Acute and Chronic Care Driver Programme - Dementias Platform

UK (DPUK) Data Portal.

• Digital Medicines Transformation Portfolio - build on the new digitalising

medicines plan from the Minister for Health and Social Services of a fully

digital prescribing approach in all care settings in Wales: the Digital Medicines

Transformation Portfolio - Digital Health and Care Wales (NHS Wales).

• Anticipated organisation of research development groups and webinars with

the other regional leads in HDR UK.

• Iteratively engage and run workshops to understand requirements with

potential users of the Phenotype Library across HDR UK and specialties

(second quarter onwards).

Key Opportunities

• Administrative Data Research (the region to focus on bringing together

administrative data and health data), eg the July workshop with the four

police forces in Wales to discuss sharing of police data for linkage with health

data; working with justice data (courts, prisons) and health data.

• ADR Wales (ADR UK).

• UKSerp - building on the expertise to host data from outside Wales (SeRP UK).

Key Challenges

• Failure to recruit researchers with appropriate skills – this could be mitigated

by extending training of the interns in the Black Internship Programme if

appropriate, as well as training of Masters students graduating from the

Health Data Science course delivered in Swansea and ailiated with HDR UK.

Partnership

Programmes

40 41

Partnership Programmes

Hubs

The UK Government’s Industrial Strategy (2017) recognised the untapped potential of the UK’s health

data as an opportunity to boost economic and health outcomes. As part of the Life Sciences Data to

early diagnosis and precision medicine challenge (D2EDPM), HDR UK was funded by UKRI to establish

the £37.5 million Digital Innovation Hub (DIH) Programme via the Industrial Strategy Challenge Fund

to develop infrastructure to support the Hubs and UK Health Data Research Alliance members. The

aim was to enhance routine NHS data and the UK’s rich cohort, registry, and research data and make it

available to industry, researchers, and innovators to use for impactful research. Following a consultation

with 1,200 across four-nations, HDR UK published a DIH Programme Prospectus in May 2019 that set

out proposals to establish three new capabilities to address challenges faced by industry and academic

researchers in accessing and using health data. These were:

• The UK Health Data Research Alliance (the Alliance): an independent, legally non-binding alliance of

health data providers, custodians and curators who develop and share standards, policies, and best

practice to demonstrate trustworthiness in health data use for research to improve human health.

• The Health Data Research Innovation Gateway (the Gateway): a web-based platform to discover and

request access to UK health datasets for research and innovation.

• Seven Health Data Research Hubs (the Hubs): UK-wide centres of excellence which focus on data

curation and create the expertise, tools, knowledge, and ways of working to maximise the insights and

innovations developed from health data.

The Hubs are very much centres of excellence with expertise and tools developing data to provide

insights. Hubs support external users from across the NHS, academia, and industry to use data for

research and innovation. Each Hub holds topic-specic knowledge, expertise and data in dierent

domains or data types. In 2021, the UKRI Medical Research Council (MRC) supported the award of

two more Hubs, Alleviate (pain) and DATAMIND (mental health), as part of a growing Hub programme.

There are currently nine Hubs that provide coverage across multiple disease areas and data sources.

Collectively, the Hubs work as a network to adopt similar ways of working and tackle challenges in data

science through being members of the wider UK Health Data Research Alliance. Below are updates from

seven of these Hubs.

Alleviate Pain Data Hub has been an active Hub for over 20 months. The core activity is to transform pain

data cohorts to the OMOP Common Data Model (CDM) and integrate with the Cohort Discovery Tool (CDT)

on the Innovation Gateway.

The Milestone 2 assessment was well received, and the Hub showcased successful PPIE across all

activities, with two PPIE leads on the project team and a pain community of more than 160 (as of August

2023). The Hub is using the expertise built from the Health Information Centre (HIC) at Dundee and the

CO-CONNECT project, with a good awareness of standards and continued development of open-source

tools to streamline data mapping in collaboration with University of Nottingham. The Hub has engaged

extensively with the Advanced Pain Discovery Platform (APDP) community with workshops, national and

international conference presentations (British Pain Society, APDP Annual Conference, NeuPSig, EFIC)

and meeting contributions (CAPE, PAINSTORM). The Hub has delivered 20 pain research cohorts on the

Gateway, knowledge sharing of the CDT across the wider pain community, an active patient group of

People with Lived Experience (PWLE) and a focal point for data strategy in the APDP.

September is pain awareness month and Alleviate will share a series of video stories from our Patient and

Public Involvement and Engagement members through their lived experience with pain. Alleviate will

also host a webinar for the wider APDP community, showcasing some of our member’s personal journeys

of living with pain. We also plan to run a UK-Wide questionnaire-based survey to gather information

around people’s experiences with Chronic Pain.

Over the next six months, Alleviate will focus on long-term sustainability. Services will be strengthened

in the coming months, with domain expertise applied to use the CDT for real-world data needs. There

will be collaborations with other pain-related or data-rich communities to solve common problems. This

will allow the Hub to facilitate wider APDP research community work to develop phenotypes for sharing

through the library. The Hub will continue to work with the Technology Team to develop the Cohort

Discovery Tool while growing the available pain relevant datasets.

Alleviate

Partnership Programmes

42 43

Partnership ProgrammesPartnership Programmes

Discover-NOW

Following the conrmation of funding by NHS England for the development of a London-wide SDE,

Discover-NOW is being scaled London-wide to create a linked primary and secondary care record for

the 10 plus million patients registered in the ve London Integrated Care Systems (ICS), creating a truly

unique SDE available for research and development. Discover-NOW also has plans to scale its operational

service oer to the whole of London. A series of stakeholder interviews, public deliberations, and cross-

workstream workshops will take place to ensure co-design of the new Discover-NOW service oer and

underpinning commercials. The Hub aims to continue our successful collaboration with partners around

AI, real world trials, retrospective analyses and population health management project, while also

exploring a self-service or direct access model.

The Hub continues to be a resourceful asset to research active institutions and commercial organisations.

To date, over 300 data access applications have been completed for Discover-NOW. Other work includes:

• Supporting high impact projects including the development of the London Asthma Decision Support

tool (LADS). This combines asthma population, clinical care, nancial and wider determinants of

health data (including air quality data) from across two London ICS. This tool is being used across NWL

to support clinical and nancial decision making for precision care planning from practice to ICB level

and has been submitted for a HSJ transformation award.

• Investigating survival and health economic outcomes in heart failure diagnosed at hospital admission

versus community settings found that diagnosis of heart failure through hospital admission continues

to dominate and is associated with signicantly greater short-term risk of mortality. This work was

published in BMJ Health and Care Informatics.

• The NWL Health Research Register, which consents patients for contact about future research

opportunities, has achieved 80,000 sign-ups, which is increasingly being used for commercial clinical

trials and is recognised as a quicker way of nding and referring patients.

Meanwhile, a priority for the business development team is to engage with current partners and

reach out to additional organisations in the health technology, pharmaceutical, consultancy, Contract

Research Organisation, and academic spaces, to not only make them aware of the potential plans and

oers of Discover-NOW, but to derive insights around the needs and requirements of these key partners

for delivering high impact health projects. The Hub has become an alliance partner with the Paddington

Life Sciences Group and have ringfenced NHSE SDE funding to create an important collaboration with

the Kings College AI Centre around imaging data. Based upon recent contracts, potential use cases

include the deployment of a medical history model to determine risk of future health complications,

patient identication and recruitment for a cardiovascular metabolic trial, retrospective analysis of a

CVD patient pathway, eectiveness of a medical device supported with AI, and a medicine proling and

optimization study.

DATAMIND

The Data Hub for Mental health INformatics research Development (DATAMIND) aims to advance mental

health research through a step change in the visibility and accessibility to NHS, administrative and

longitudinal data. This will be achieved through greater discoverability, annotation, harmonisation, and

advanced analytics by ensuring the inclusion of diverse groups of under-represented individuals with the

greatest clinical need.

The Hub has focused on four core activity areas and has also identied four challenge areas that have

guided research. These are: Children and Young People, Excluded and Underrepresented Groups,

Interfaces between Physical and Mental Health, and Severe Mental Illness (SMI). Specically building

on the MRC Pathnder community, the Hub has also been working on six identied Road Builder

Innovations which are being delivered at dierent Milestones over the course of the project. These will

potentially become Hub Core Activities over time, working across and extending further into Challenge

Areas. The Road Builder Innovations are:

• Discoverable Schools.

• Discoverable excluded and under-served groups.

• Linking Physical and mental health in SMI.

• Widening availability of mental health care text analytics capabilities.

• Digitally Enhanced Trials.

• Drug discoverability.

To date, the DATAMIND team has focused on workshops and the identication of training needs for

early career researchers, followed by provision of courses. The data literacy course has been developed

alongside our PPIE group to develop the capacity of people with lived and living experience of mental

health in Mental Health Data Science research. This online course will be released soon. There is an

extensive range of tools being developed to support a range of research needs: Core Mental Health

DataSet (CMHDS) collection tool; equity audit tool for clinical trials; Mental Health Text Analytics Cloud

(MH-TAC); VELA tool to simplify information extraction from large, linkable, mental health data sources;

and PHENOMIND for making phenotype library code lists and algorithms available for research data.

The partnership with the National Institute for Health Research Clinical Research Network (NIHR CRN) is

important for widespread uptake.

There are eorts to integrate the Catalogue of Mental Health Measures and Landscaping International

Longitudinal Data project with the Innovation Gateway. The Welsh work on hard-to-reach groups shows

real value of dierent types of data assets, and it is good to see that the Natural Language Processing

work is being built out. The Hub established an Industry Forum to connect DATAMIND with several

industries, including the pharmaceutical sector, to explore the possibility of developing work that would

address mental health needs. During these discussions, various gaps in the eld have been identied

such as the possibility of using genomic data collected as part of a pharma-sponsored clinical trial to

understand how antidepressants work and why they work better for some people than for others. This

also led to the application of a Wellcome Trust call which was awarded in December 2022, contributing to

the sustainability of DATAMIND. Overall, the Hub is showing excellent progress and integration with other

work/investments.

Over the next six months, the Hub will continue to expand its online presence with its website. Two key

additions that will be added to the DATAMIND website are an online data literacy course developed in

collaboration with McPin, DATAMIND, and the Super Research Advisory Group (SRAG), covering topics

like patient rights, research, and risks. Additionally, the DATAMIND glossary, created with assistance

from the PPIE team, will provide easily understandable explanations of essential terms in mental health

research and data analysis, fostering eective communication and collaboration among stakeholders.

The DATAMIND PPIE group will continue their involvement in co-producing and developing guidelines

and strategies related to privacy and data sharing and other ongoing documents to aid the eld of mental

health and its community, including industry. The next Industry Forum will be held late in 2023 with PPIE

for them to engage directly with industry partners. The next data science events are also in preparation

for late 2023 and will incorporate the feedback received from early career researchers and attendees

of previous workshops. They will also focus on working with HDR UK Futures and the MQ Mental

Health Research charity for engagement opportunities and to develop training modules for Continuing

Professional Development-style-learning. Work continues to onboard datasets through the Gateway.

44 45

Partnership Programmes

PIONEER

At PIONEER, access to detailed, individually linked health data, supported by expert consultancy and

highly trained sta, is transforming healthcare for our population. The PIONEER Hub continues to

facilitate groundbreaking research and innovation in acute care. By providing researchers with access to

comprehensive and diverse datasets, the Hub has enabled discoveries that drive the development of new

treatment modalities, care pathways and medical technologies. There is now a scalable, interoperable

system that can be federated to deliver high agility regionally, nationally and internationally. PIONEER is

a model that the team designed and own.

PIONEER has achieved remarkable success by serving over 80 requests across eight sectors, including

four from industry partners. These achievements have been recognised and supported by prestigious

funders, resulting in grant awards totalling >£29 million. PIONEER has also made signicant

contributions to data accessibility, oering 45 platinum-rated datasets through the HDR UK gateway.

Moreover, the Hub has secured funding to ensure its nancial sustainability for the next two years,

ensuring continued innovation and support for the healthcare community.

PIONEER has always been actively engaged in collaboration with numerous organisations, as well as

patients and public. Over the past six months they have focussed on partnerships established via the

HDR UK Alliance, namely the Regional Delivery Group working to deliver impactful research across

the country. PIONEER has continued to participate in several large multi-centre grants and projects,

addressing crucial areas such as multi-morbidity, polypharmacy, and the DARE TRE-FX grant, which

explores technical solutions to the federation of analytics. They have continued to consult with patients

and the public on a number of health research projects, including co-production of a patient leaet for

four acute hospitals, as well as a script of a patient-facing video explaining federated analysis. The next

six months will see PIONEER initiating studies as part of the HDR UK second ve years Medicines in Acute

and Chronic Care driver enabling further collaboration across primary and secondary care.

Furthermore, the team are working on projects with the Patient Safety Research Collaborative that

brings together NHS trusts, universities and private business to evaluate how digital tools can support

clinical decision-making and reduce risks of harm for expectant mums and anyone in need of emergency

treatment. They are also carrying out work on the Acute Care theme for the NIHR Birmingham

Biomedical Research Centre, which combines world-class strengths in immunology and inammation

research and extensive experimental medicine infrastructure. The next six months promise an exciting

continuation of collaborative eorts to advance acute care and improve patient outcomes.

Partnership Programmes

NHS DigiTrials is a health data hub alumnus continuing to go from strength to strength and has recently

released a new and improved version of its feasibility tool. This includes a range of accessibility and

usability enhancements which have been tested with users. The NHS DigiTrials Feasibility Self Service

tool enables researchers to determine if there are enough suitable people for their clinical trials in

a matter of minutes. Researchers can run queries using lters such as disease diagnoses, hospital

interventions, and primary care medicines, within a secure environment. The Hub continues to develop

our other services to help trialists communicate with their participants and use routine healthcare data

to follow-up their cohort.

If you would like to hear more or think the Hub can help you with your research you can email the team

at enquiries@nhsdigital.nhs.uk. In the meantime have a look at their blog with more detail of what the

tool can do: Clinical trials get faster - NHS Digital.

NHS DigiTrials

INSIGHT

INSIGHT, the Health Data Research Hub for Eye Health, is the world’s largest ophthalmic bioresource,

with over 25 million retinal images and associated clinical data. An NHS initiative led by Moorelds Eye

Hospital in partnership with University Hospitals Birmingham NHS Foundation Trust, INSIGHT makes

routinely collected eye data available for approved research to advance development of new healthcare

treatments and diagnostic technologies for eye disease and systemic diseases such as diabetes and

dementia. The INSIGHT team and its partners are applying advanced analytics, including machine

learning and articial intelligence, to the Hub’s datasets for patient benet. New data from Moorelds

Eye Hospital and University Hospitals Birmingham is added on an ongoing basis. For more information

visit the INSIGHT website and follow INSIGHT on Twitter @INSIGHTeyehub for updates.

Gut Reaction

Gut Reaction is a unique, secure data resource designed to facilitate academic and industry research in

Inammatory Bowel Disease (IBD), working with the IBD community to improve treatment options and

patient outcomes through safe, transparent and responsible use of patient data.

Gut Reaction is the world’s largest repository of high quality, consented and linkable data supporting

research to change the lives of people living with IBD. The goal during the initial funding period was

to integrate health and research data from patients recruited to the NIHR IBD BioResource who have

consented for information from their health records to be held in a research database. The specic

aims are:

1. To transfer health records, including digital pathology and digital imaging data, from NHS Trusts, and

HES data from NHS Digital, to AIMES, where clinical and phenotypic data are currently held.

2. To integrate these data with genomic data held at the University of Cambridge High Performance

Computing (HPC) Service in a secure environment in Microso Azure, where they can be analysed

anonymously, with the capability to re-identify to allow contact and recall of individuals.

The Hub has benetted from industry interactions (e.g. GSK and AstraZeneca). AstraZeneca wish to

conduct a fully funded data refresh across the whole BioResource, in exchange for data access. Gut

Reaction has continued its work as part of the NIHR Bioresource. Working with partners, collaborators

and the IBD community, they will continue to support research that makes a dierence to those living

with IBD. Gut Reaction are currently reviewing how data is housed, together with the need to supply

this under licence to various requestors in dierent environments. There have been plenty of learnings

from UK BioBank and the expectation is that there will be a large programme of work over the next one

to two years.

46 47

Partnership Programmes

BREATHE was a unique collaboration driving the use of health data in research to transform

respiratory health.

BREATHE’s legacy lives on through the establishment of three respiratory data registries in England,

Scotland and Wales. The groundwork to create a fourth in Northern Ireland is underway, tying in with

the activities of the Inammation and Immunity Driver Programme. In each of these nations, BREATHE

collaborated with Trusted Research Environments (TREs) and data providers to create cohorts of patients

with chronic respiratory diseases, specically: asthma, Chronic Obstructive Pulmonary Disease (COPD)

and Interstitial Lung Disease (ILD). BREATHE worked with the Clinical Practice Research Datalink (CPRD)

in England, DataLoch in Scotland, and the Secure Anonymised Information Linkage (SAIL) Databank in

Wales. The Northern Ireland registry will be built in collaboration with the Honest Broker Service and

Queen’s University Belfast.

All registries provide a baseline harmonised set of criteria and clinical coding for how asthma, COPD,

and ILD should be characterised in routine health records, contain research-ready data related to

patients’ demographics, diagnoses, condition events (e.g. ongoing GP or hospital care) and medications,

and also provide the facility to link to wider records pertaining to comorbidities, other conditions, and

healthcare history.

The DataLoch Respiratory Registry, currently covering South-East Scotland residents within the NHS

Lothian catchment, holds additional information on Cystic Fibrosis and Wheeze (a common respiratory

symptom). In England and Wales, the processes of linking further conditions to the registries is

relatively straightforward, subject to governance and dependent on the inclusion criteria of specic

research studies.The registries have been expertly curated with clinical input and are harmonised as far

as possible (across multiple clinical coding sets and agreed individual-level characteristics), making it

easier to conduct pan-UK analyses. A paper detailing the methodology used in the cohort construction

and harmonisation processes has been submitted for publication. Once published, the curated script and

codes used to create the asthma, COPD and ILD cohorts will be made available, in collaboration with the

team who created the cohorts. This will allow for equivalent cohorts to be curated elsewhere, enabling

curation and linkage within and to wider records held by the data custodians that control them.

Another, separate paper is also in progress. Using data from England, Scotland and Wales, this

epidemiology study aims to generate baseline prevalence and incidence of asthma, COPD and ILD across

the UK, within the context of the curated cohorts. It will also include asthma data from Northern Ireland.

BREATHE

Partnership Programmes

DATA-CAN continues to work towards its strategic ambition of facilitating access to high quality cancer

data from patients across the four devolved nations of the UK. The Hub continues to work with, and

support, clinical, academic, charitable and commercial partners, oering expert advice, guidance

and analytic expertise on routine datasets collected nationally or within individual cancer centres. In

partnership with NHS England, DATA-CAN had helped develop an SDE containing both cancer-specic

and general healthcare datasets on all cancer patients diagnosed in England. Current work to evaluate

this data is underway through an evaluation of the impact of the COVID pandemic on cancer referrals,

diagnosis, treatment and outcomes. This follows up on their previous work on the initial impact of the

pandemic on cancer services and cancer patients, where they were the rst to highlight the disastrous

impact of the pandemic on cancer in the UK. This work was the catalyst for a pan-European study that

showed that 100 million screening tests were not performed and up to one million cancer diagnoses

may have been missed due to the negative impact of the pandemic.

DATA-CAN works with cancer centres across the UK to identify cohorts of patients with a specic

diagnosis to describe treatment pathways and outcomes. This work, oen performed for commercial

partners, supports a deeper understanding of routine care or provides real-world evidence to support

funding approvals for new cancer therapies within the UK. DATA-CAN continues to expand this

network actively seeking new partners to enhance the value of this oer. DATA-CAN aims to deliver

this work through the adoption of the OMOP common data model and is evaluating privacy preserving

technologies including federated analysis to further enhance this work. DATA-CAN continues to

work with a range of commercial partners to support the evaluation and implementation of novel

technologies aiming to enhance the collection, curation and use of routine health care data in this

work. This includes work linking structured data to unstructured letters and reports, digital imaging and

pathology, detailed molecular and genetic phenotype and most recently patient reported outcome data.

Our PPIE representatives remain critical to DATA-CAN’s mission and continue to guide and inuence our

strategy ensuring that the work performed delivers value to patients and the NHS more widely. DATA-CAN

received the HDR UK Impact of the year award for its work on evaluating continuous versus intermittent

cetuximab treatment for colorectal cancer. The evidence that this work generated was critical to NHS

England approving a policy change to allow treatment breaks for patients with colorectal cancer.

DATA-CAN will work closely over the next ve years with HDR UK’s Big Data in Complex Diseases Driver

Programme to maximise the use of the national-scale datasets highlighted above, aiming to enhance

early diagnosis, better treatment and improved outcomes.

DATA-CAN

48 49

Partnership Programmes

DARE UK

The DARE UK programme is funded by UKRI as part of its Digital Research Infrastructure portfolio of

investments, which supports the development of a coordinated vision for digital research infrastructure

in the UK. DARE UK is a pan-UKRI, cross-domain programme - its scope covers all types of sensitive data,

including data about education, health, the environment and much more. There is growing consensus

(including from Phase 1 DARE UK recommendations, the Goldacre Review, the UK Health Data Research

Alliance TRE Green paper, and the DHSC Data Saves Lives policy paper) that all sensitive data should

only ever be accessed and analysed by researchers within a TRE. Central to the DARE UK programme’s

ambition is to enable the development of a national interoperable network of secure digital research

infrastructures or TREs, laying the foundation for a next-generation ecosystem of TREs for advanced data

research for the public good, using a cloud-rst design philosophy.

Delivery of Phase 1 of the programme began in July 2021 and is being delivered with joint oversight

from HDR UK and ADR UK. Phase 1 is an extensive programme of community engagement that is an

essential foundation for developing a clear vision of the needs of dierent research communities, and to

address the interests and concerns of the public around the use of sensitive data for research. It will lead

to a community co-designed blueprint for an interoperable network of next generation TREs alongside

a model for delivery of Phase 2 which will begin to iteratively build, test, and establish this vision. As

Phase 2 approaches, with several key outputs from Phase 1 in development such as the initial version

of the Federated Architecture Blueprint and the outputs of the portfolio of Driver Projects, the DARE UK

programme is considering how best to deliver the programme’s ambitions in collaboration with the

various sensitive data research communities and stakeholders, while incorporating the full range of

earlier outputs of Phase 1 (for example the Phase 1 DARE UK recommendations, the portfolio of Sprint

Exemplar Projects, and DARE UK Public Dialogue amongst others).

Important for this community to consider are the kinds of inter-disciplinary, cross-domain scientic

use cases that could be realised through a national interoperable network of secure TREs and to

collaboratively begin to consider the opportunities for co-delivery of components of the DARE UK

vision - for example through involvement with DARE UK Community Groups - in the lead up to Phase 2

of the programme.

BHF Data Science Centre

The BHF Data Science Centre (BHF DSC) is a partnership between HDR UK and the British Heart

Foundation and sits within HDR UK.

We work with a wide range of partners including patients, public, clinicians, researchers and NHS

organisations to help them carry out research using health data into the causes, prevention and

treatment of all diseases of the heart and circulation (such heart attacks, stroke and vascular dementia).

We do this to ensure new advances in treatment and care for diseases of the heart and circulation get to

the patient as quickly as possible. The centre started in January 2020 and has initial investment from the

BHF for £10 million over the rst ve years. The BHF DSC has the following thematic areas:

• Structured data - this work is driven by the CVD-COVID-UK/COVID IMPACT Consortium which has enabled

an England-wide electronic health record (EHR) resource within the NHS England SDE.

• Unstructured Data.

• Personal Monitoring Data.

• Computable Phenotypes.

• Data enabled clinical trials.

• Enhancing cohorts.

• Diabetes Data Science Catalyst.

Community-wide opportunities include joining the CVD-COVID-UK/COVID IMPACT Consortium to enable

rapid access to NHSE SDE data and national TREs in Scotland and Wales, collaborative opportunities with

BHF DSC and SAIL Databank to enhance cohort study data by linking to EHR.

Partnership Programmes